JP2923648B2 - Method and apparatus for generating color characteristics of an object - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to a color image generator for computer image generation (CIG).
Preliminary processing to find the appropriate color information for the cell fabric, or more specifically, the display, is a very common red, green, blue (RGB), or red used in other CIG schemes. , Green, blue, translucent (RGBT) color space, rather than a color cell fabric for real-time CIG devices such as those performed in the YIQ color space.

For example, in pending U.S. patent application Ser.No. 943,690 filed Dec. 19, 1986, a full color cell fabric generator useful for CIG devices is provided for each of the RGBT components in the high resolution display channel of the CIG device. On the other hand, it requires 16 parallel paths to calculate the dough. Even if the cell map data is compressed before performing the parallel calculation operation, 64 parallel logical paths per high-resolution channel still remain.
The associated hardware and associated dimensions and costs are such that this color cell fabric scheme cannot be used in CIG devices, and therefore only black and white fabric can be used.

Generally, human eyes are less sensitive to color changes than to luminance changes. Taking advantage of the fact that the eyes are relatively sensitive to luminance and relatively insensitive to color changes, the computational load of the CIG device that presents the color cell fabric of the image and the necessary hardware are reduced. While decreasing, it is desirable to maintain the proper color rendering of the image so that the observer has the appropriate color information and training clues.

Typically, CIG devices can be used to train operators / observers to perform tasks such as flying an aircraft. The displayed image is not necessarily as expected from the video of the scene created by the camera.
Although it is not necessary to have a sense of reality, it is expected that the trainee makes a conditional reflection as a reaction to the simulated scene and performs the same reaction in response to the same external environment as the simulated one. In order to do so, there must be enough realism.

In the present invention, one potential possibility as a practical solution to the above mentioned problem is to use the YIQ scheme for color cell fabric processing. In the YIQ method,
Y represents luminance or brightness, and I and Q in phase and quadrature respectively represent chroma.

Color television devices use MIQ video, where M represents luminance and I and Q represent the chroma of the image to be displayed, but generally the processing is performed using analog or continuous signals, And no special consideration for negative signal values. In contrast, in computer image generators, processing is typically performed in the digital or discrete domain, where accuracy is limited to a predetermined number of bits and functions with negative values can be efficiently implemented. To process well,
Data modification may be used.

Accordingly, it is an object of the present invention to reduce the hardware and logic circuitry required to process color cell fabric information in a computer image generator.

Another object is to provide an apparatus and method for using YIQ color cell fabric processing in a computer image generator.

SUMMARY OF THE INVENTION In the present invention, red, green and blue (RGB) original image descriptors are converted to corresponding luminance (brightness), in-phase and quadrature (YIQ) forms. In one aspect of the invention, RGB data, such as those available from a database or cell fabric map, is converted to YIQ data, and then the YIQ data is stored in another database or cell fabric map, and the map is used to process the image. Used as a database for RGB
The conversion from data to YIQ data is only a change in the usage that defines the attributes of the color cell fabric of the object, and the conversion can be performed efficiently. Of course, the database or color cell fabric information can be limited directly to the YIQ format without generating or recalling the corresponding RGB data.

To achieve another advantage of the present invention, the YIQ data expressed in full scale units is rescaled (ie, positive for negative values of I and / or Q) before processing. ), Thus allowing the values of Y, I and Q to be represented by a smaller number of bits, thus requiring less parallel bit transmission paths for processing. Furthermore, since the human eye is less sensitive to color changes than brightness changes, the value of Y representing brightness is calculated at the resolution of ordinary cell fabric, i.e. at a certain level of fineness. For (representing chroma or color)
The values of I and Q are 1/4 or 1/1 of the resolution of ordinary cell fabric.
Calculations at a reduced cell fabric resolution, such as 6, or at a coarser granularity level can save further hardware.

For devices that process RGB data, for each of the R, G and B components, typically using a 16-bit resolution and the corresponding 16 parallel data paths, the total number of paths is 48, An apparatus for processing scaled and reselected YIQ data according to the present invention, such as processing I and Q at 1/4 of the resolution of Y, requires 16 parallel data paths for Y. However, for each of I and Q, only four parallel data paths are required. As a result of a total of 24 parallel data paths when performing YIQ processing according to the present invention, a 2: 1 saving is achieved as compared with the RGB method. The color of the final displayed object is expected to be slightly lower when using YIQ processing than that obtained by RGB processing, but hardware savings when using YIQ processing according to the present invention. And the associated cost savings, especially in devices where close color correspondence with the actual object is not considered important.
There is much more to compensate for this drop. Furthermore, according to a subjective test,
It has been found that there is almost no perceptible difference between the conventional RGBT processing and the YIQT processing of the present invention.

After processing, the resulting Y, I and Q values, expressed in scaled and reselected units, are converted to the original full scale Y, I with the help of a table lookup, etc.
And return the magnification to Q units. Full scale Y, I and Q values, typically in digital form, can be converted to RGB values. The RGB values thus obtained can be further processed and supplied to a conditioning circuit before final use in a display circuit.

The translucency data is part of an RGBT database or a color cell texture map. In one embodiment of the present invention,
The translucency data from the RGBT database can be processed in the same way as the Y data component of the YIQ data extracted from the RGBT database. In another embodiment, the YIQ data is processed as described above, and the value of the Y data is scaled back to the original full-scale Y unit before using it to access the table lookup. Do. As a result, a signal having a translucency value in units of full scale is obtained. This signal, representing translucency, is typically in digital form, but, like the translucency Y data, can be provided to circuitry for further processing and conditioning before final use in the display circuit.

The novel features of the present invention are specifically described in the appended claims, but the structure, operation, and other objects and advantages of the present invention will be best understood from the following detailed description of the drawings. Let's do it.

DETAILED DESCRIPTION In one color fabric generation system, such as that illustrated in pending U.S. patent application Ser. No. 943,690 filed Dec. 19, 1986, in order to limit the color space, red and green are orthogonal to each other. And blue (RGB) color axes. Another property, called translucency (T), which describes the degree of light transmission through an object, which can vary from none (opaque) to full (transparent), can be processed in the same way with RGB data. I can do it.

As an example of how to use the invention, a description will be given of the case where the invention is used in a CIG device as exemplified in pending U.S. patent application Ser. No. 865,591 filed May 21, 1986. . To provide the necessary information for cell mixing and cell smoothing as shown in this U.S. patent application, the full color cell fabric is a high resolution channel of the CIG device, each of the four RGBT components. On the other hand, 16 parallel passages for calculating the dough are used.

Although the same reference numerals are used for similar parts throughout the drawings, FIG. 1 shows the RGB orthogonal system and the YIQ of the present invention.
The relationship between orthogonal systems is shown. Y in this specification
The component represents brightness (luminance), and the I and Q components represent chroma of a chrominance signal. The I and Q portions may be referred to as in-phase and quadrature components. As will be described in detail later, when the Y component is represented by the same pixel resolution as that used in the processing in the corresponding RGB system, while the I and Q components are represented by the lower pixel resolutions respectively, The advantages of the present invention are obtained.

As shown, the RGB system 10 can be represented by three mutually orthogonal axes labeled R, G, and B, respectively. Each axis of the orthogonal triaxial system 10 represents the monotonically varying color intensity of the corresponding RGB color from the origin of the system. A color that is not a pure color rendering of one axis color is assigned a position within the three-axis system that represents the respective intensity combination of the red, green, and blue primaries. Typically, each axis 256, i.e., is divided into two ^eight discrete intensity, the origin (0,0,0) represents black, points (255, 255, 255)
Represents white.

Similarly, the YIQ system 20 can be represented by three mutually orthogonal axes labeled Y, I and Q, respectively. The Y-axis represents a monotonically changing brightness or luminance characteristic, and the I and Q axes each represent monotonically changing chroma information. One way to determine the orientation of the YIQ system, especially from an existing RGB system such as that used to construct a database, is to set the Y axis at the point (255,255,255) from the RGB system origin (0,0,0). Or, the same R, G of TGB system
And to match the vector to any other point with the B and B coordinates to maintain the orthogonality and relative position of the Y, I and Q axes. With such a conversion, RGB
The coordinates of one point in the system can be mapped to the corresponding point in the YIQ system according to the following equation:

Y = 0.299R + 0.587G + 0.114B (1) I = 0.596R-0.274G-0.322B (2) Q = 0.211R-0.523G + 0.312B (3) Here, Y, I and Q in the RGB system Represents the coordinates of a point in the YIQ system corresponding to a point having coordinates R, G and B. If the value along each axis in the RGB system is in the range of 0 to 255 (ie, 256 bits), then Y is in the range of 0 to 255, I is in the range of about -151 to +151, and Q is in the range of about -151. The range is from 133 to +133. RGB
Each point in the database can be converted to a corresponding point in the YIQ system to create a YIQ database. Of course, the Y, I, and Q coordinates can be directly specified and stored in the database, as in the case of creating the RGB database, without determining the corresponding R, G, and B coordinates. Furthermore, RG
Since both the B and YIQ systems represent usages that determine data points, the conversion from one system to the other does not require extensive calculations. Although it is possible to directly manipulate the Y, I and Q coordinates obtained from equations (1), (2) and (3), the resulting Y, I and Q coordinates are scaled. ,
It has been found more convenient to convert the I and Q values to positive numbers if necessary.

FIG. 2 shows an apparatus useful for converting RGB data into YIQ information in accordance with the present invention. An RGBT database or cell texture map 30 has scene descriptors stored on a bulk storage medium such as a magnetic disk or tape. A device that uses an RGBT database with associated control functions is described in U.S. Patent Application Serial No.
No. 943,690. It should be appreciated that the apparatus shown in the figures can be used equally well for operations using RGB databases, such as those that do not require translucency related components.

When the RGBT data of the database 30 is needed to determine the characteristics of the pixels to be displayed on the display device (not shown), the descriptors of the R, G, and B data points are respectively stored in the database 30. The descriptor of the associated T data point is made available from another output of the database 30. As is well known, the database 30 requests RGBT data in response to timing and synchronization signals (not shown) from the CIG device.
A conversion circuit 35, whose input is coupled to the output of the database 30 for receiving R, G and B data, responds to the supplied R, G and B data in accordance with equations (1), (2) and ( According to 3), generate corresponding Y, I and Q coordinates. As a result, the Y, I, and Q coordinate information that can be used for the output of the
It is supplied to the input of a magnification circuit 38. The translucency information T can be delayed so as to compensate for the internal delay of the conversion circuit 35, which is coupled from another output of the database 30 to another input of the magnification circuit 38, so that the magnification circuit The YIQ data received by 38 corresponds to the RGBT data supplied from database 30. Y, I, Q and T supplied with the magnification circuit 38
Each of the information is multiplied by a factor, and if the value of I and / or Q is initially negative, it is re-selected as a positive value. The resulting multiplied and multiplied values, and the re-selected values, and (or Y ', I', Q 'and T') are made available to the output of the magnification circuit 38, respectively. Timing and synchronization signals (not shown) convert YIQ data and Y'I '
To adjust the storage of the Q'T 'data, it is supplied from a synchronization source (not shown) to a database 30, a conversion circuit 35 and a magnification circuit 38. Conversion circuit 35 and magnification circuit 38
Is conveniently configured as a part of a computer program, as is well known.

In the presently preferred embodiment, conversion circuit 35 and magnification circuit 38 are used to generate the scaled Y ', I', Q ', and T' data which are available to the respective outputs of magnification circuit 38. Work offline. These data are Y'I '
It is supplied to the Q'T 'database or cell fabric map 40 and stored therein. The database 40 contains Y ',
It has respective outputs for the I ', Q', and T 'data, but can be connected to provide Y', I ', Q', and T 'data to the processing circuitry, as described below. I can do it.

A scaling circuit 38 determines the maximum and minimum values for each of the Y, I and Q cell texture signals received from the conversion circuit 35 throughout the cell texture map. Each of the Y ', I' and Q 'fabric cells in the Y'I'Q'T'cell fabric map 40 is assigned a preferably 4-bit compressed value representing the value of the compressed fabric cell. . Y ', I' and Q '
Are determined based on the relative positions of the uncompressed fabric signals, respectively, between the minimum and maximum values of the corresponding uncompressed cell fabric values of Y, I, and Q. . Using 4 bits to compress the value of the fabric cell, the reassigned minimum can be represented by 0 and the reassigned maximum can be represented by 15. By first sampling the original values over the values of the Y, I and Q cell dough to determine the minimum and maximum values, then assigning the compressed Y ', I' and Q 'values Since the entire available range of compressed values is used, an efficient allocation of available compressed values is guaranteed.

Although the connections between the circuit components are generally indicated by a single line, it will be appreciated that parallel data transfer may be used, if desired, to increase the overall throughput of the device. Further, separate connections and circuits may be used as desired for each of the R, G, B, and T components and the Y, I, Q, and T components of data.

The database 40 responds to signal commands (not shown in the drawings) relating to the portion of the scene to be displayed that is to be processed, at the appropriate level of fineness or resolution, with Y ', I', Q 'and Supply T 'data to each output. For example, each of the R, G, B, and T components and each of the Y, I, and Q components of the converted YIQ data can use RGBT data represented by a 16-bit resolution. Each of the corresponding scaled Y ', I', Q ', and T' data can be represented with a 4-bit resolution.

Further, RGBT information can be stored at different levels of fineness (LOD). The resolution of a certain LOD corresponds to a predetermined distance section from a viewpoint to a scene to be displayed. The intended area, i.e., the smallest unit cell or pixel for defining the object, and thus the finest LOD, such as the finest resolution, is closest to the viewpoint and Each coarser LOD has a larger cell, correspondingly less fineness and coarser resolution. Scheduled LO
RGBT information for object D is generally stored in database 30
In the place of the scheduled LOD. But
To achieve the advantages of the present invention, the Y 'data corresponding to the RGBT data is stored in the database 40 in the database 30.
The associated I 'and Q' data is stored in the database 40 at the coarser LOD, typically at the next coarser LOD. Details of the storage of Y, I, and Q will be described later with reference to FIGS. Further, if desired, Y ', I', Q '
And a separate database (not shown in the figure) may be used to store each of T. and T '.

FIG. 3 shows an apparatus for processing YIQT color information. For a detailed description of the smoothing and mixing operations, see the above-cited pending US Patent Application No. 865,591. Here, a description will be given of processing within a range necessary for understanding the present invention. Two multiplicity values in resolution (LOD) or resolution Y 'from database 40 (FIG. 2)
Are supplied to respective inputs of the LOD (N) Y 'smoothing circuit 50 and the LOD (N + M) Y' smoothing circuit 52. Similarly, the multiplied and / or reselected values I 'and Q' and the multiplied value T 'from the database 40 (FIG. 2) are the LOD (N) And LOD (N + M) I 'smoothing circuit 54, LOD (N) and LOD (N + M) Q' smoothing circuit 56
And LOD (N) and LOD (N + M) T 'are supplied to respective inputs of the smoothing circuit 58. To avoid unnecessary repetition, I ',
For the Q 'and T' channels, only one smoothing circuit is shown. It should be appreciated that the circuitry for the I ', Q', and T 'channels can be configured similarly to the circuitry for the Y' channel. The subscripts N and N + M represent the respective levels of fineness.

The outputs of the Y 'smoothing circuit 50 and the Y' smoothing circuit 52 are respectively
Smoothed values of LOD (N) and LOD (N + M), _Y'SN and Y _{'S (N + M)} , may be utilized, which are coupled to respective inputs of Y 'mixing circuit 60. A mixing circuit 60 proportionally mixes the values of the _Y'SN signal and the Y _{'S (N + M)} signal. Y _'SN signal and _{Y' S (N + M)} Y represents a mixed value of the signal _'B signal Y' out to an output of the mixing circuit 60. The output of the Y 'mixing circuit 60 is connected to the input of the magnification means 70. Acting on mixing Y _B signal magnification means 70 came supplied. This is expressed in units multiplied by the magnification, thus the way to or convert such changing magnification units of the same Y signal as obtained values of Y _B signals from equation (1). At the output of the magnification means 70, a Y signal remultiplied is output.

Similarly, the output of the I 'smoothing circuit 54 is connected to the input of the I' mixing circuit 62, and the output of the I 'mixing circuit 62 is connected to the input of the magnification circuit 72. Similarly, the output of the Q 'smoothing circuit 56 is Q'
The output of the Q 'mixing circuit 64 is connected to the input of the magnification circuit 74. At the outputs of the magnification circuits 72 and 74, I and Q signals which have been remultiplied are output. Besides redial the magnification value of the came is supplied I _B and Q _B signals, the multiplying circuit 72 and 74 respectively, where appropriate, assigns a negative value to a value that call again the magnification.

When processing the separate translucency signals, a scaled translucency signal T 'is provided to the input of a T' smoothing circuit 58 for LOD (N) and LOD (N + M). Smoothing circuit 58, T '
The processing of the translucency data by the mixing circuit 66 and the magnification circuit 76 is performed by the Y ′ smoothing circuit 50 for the scaled Y ′ data,
The processing is the same as that performed by the Y 'smoothing circuit 52, the Y' mixing circuit 60, and the magnification circuit 70. The translucent T signal remultiplied by the magnification output from the output of the magnification circuit 76 is expressed in a digital form (ie, a discrete form) in the same unit as the translucency signal of the RGBT database 30, and is output to the output of the magnification circuit 76. .

The Y, I and Q signals from the magnification circuits 70, 72 and 74 are supplied to the converter 80. The converter 80 has the formula (1), (2) and (3)
R = Y + 0.956I + 0.621Q (4) G = Y−0.272I−0.647Q (5) B = Y−1.106I + I.703Q (6) From the Q information, the values of R, G and B are determined. The value of the composite RGB data is generated in the required format at each output of the converter 80, and is post-processed and conditioned with the appropriate translucency signal T from the scaling circuits 76, 78 before being quoted before. The final display is made as described in detail in the two U.S. patent applications.

When encoding the translucency information together with the Y signal component, a T magnification circuit 78 is used. This input is 'connected to the output of the mixing circuit 60, Y' Y receive _B signal. A T magnification circuit 78 generates a T signal in which the output is multiplied again. When translucent information is encoded in this manner, the components of the translucent channel, ie, the T 'smoothing circuit 58, the T' mixing circuit 66, and the magnification circuit 76 are not required. Each of the Y magnification circuit 70, the I magnification circuit 72, the Q magnification circuit 74, and the T magnification circuits 76 and 78 are in a table lookup, and the input receives the respective mixed signal supplied thereto. It can be like. Each input signal is used to address a look-up table, and in response, a respective one of the digital signals is obtained.

In one embodiment of the present invention, Y ', I', Q ', and T' can all be represented by 4-bit compressed data values. By processing in the corresponding smoothing and mixing circuits, an additional 4 bits of fractional data are added to the value of each resulting mixed Y ', I', Q 'and T' signal, which is respectively Is supplied to the multiplication circuit. As a result, using the value of the compressed data of 8 bits as a whole, the corresponding magnification circuit
70, 72, 74, 76, 78 lookup tables can be addressed. The output of the look-up table is the output of the scaling circuit, which is the uncompressed Y, I, Q and T corresponding to the values of the compressed data used to address the look-up table. Assigned in a scheduled manner based on the value of the data.

FIG. 4 shows a scheduled area 90 and a scheduled area 90 according to the invention.
The Y data storage for LOD is shown schematically.
The value stored in each cell or pixel location is represented by Y with two subscripts, such as Y _mn . m represents a row and n represents a column. The translucency data T when not encoded in the luminance Y data is stored at the same resolution as the luminance Y data.

FIG. 5 schematically shows the storage of I data corresponding to the Y data of FIG. 4 for the same scheduled area 90. The value stored in each cell or pixel location is represented by I with two indices, such as I _xy. X
Represents a row, and y represents a column. A data storage scheme similar to that shown in FIG. 5 can be used to store Q data corresponding to the Y data in FIG.

Note that one cell 94 of the I data (and thus also the Q data) of FIG. 5 covers four times the area of one cell 91 of the Y data of FIG. In other words, YIQ
In the data, the Y data point is stored at the scheduled LOD,
The corresponding I and Q data points are each stored at the next coarser LOD. Therefore, four cells 9 of Y data (FIG. 5)
The data limiting 1,93,95,97 are the coordinates Y ₁₁ I ₁₁ Q ₁₁ , Y
_It is represented by data from cells having ₁₂ I ₁₁ Q ₁₁ , Y ₂₁ I ₁₁ Q ₁₁ and Y ₂₂ I ₁₁ Q ₁₁ . That is, each of the four Y coordinates is associated with the same pair of I and Q coordinates. If desired, for the corresponding luminance Y, the I and Q data can be stored at other coarser LODs.

In one aspect of the present invention, a magnification circuit 38 (FIG. 2) receives Y, I, Q and T data, each typically represented in digital form, having a resolution of 16 bits, and outputs respective data. To make the data Y ', I', Q 'and T' available. The Y, I, G and T data supplied to the scaling circuit 38 are scaled or mapped to a range of 0 to 15 (ie, which can be represented by 4 bits of resolution), and if necessary, I and And / or the Q data is re-selected to a positive value. The translucency data T is scaled according to the same logic as the luminance data Y. The intermediate values of the Y, I, Q, and T data (ie, between the minimum and maximum values) are assumed to be uniformly spaced between the minimum and maximum values (ie, a first order transform): It is mapped in the range of 0 to 15.

When using a 16-bit resolution for each of the R, G, B and T data, parallel processing requires a total of 64 lines. The corresponding Y, I, Q and T data are scaled, and each scaled value Y ', I', Q 'and T' is represented by a 4-bit resolution, so that 24 lines are used for parallel processing. Only wires are needed. Do not process the translucency T data independently.
As described in the figure, when deriving from the luminance data Y,
Further hardware savings are possible.

When using the 4-bit resolution described above for each of the Y, I, and Q color components, each YIQ cell has an average of 6 bits, ie, 6 bits, to completely define the color in scaled units, ie, It requires 4 bits for luminance Y data and 1/4 of the total 8 bits required for I and Q chroma data.

Returning to FIG. 3 again, an example of the sample processing operation will be described. The operation is performed on an area called a span in the image to be displayed. One type of span includes eight pixels in a row or line and has a length of eight lines. A pixel is the smallest addressable element of a display device (not shown).

Pending U.S. Patent Application Serial No. 865,59, cited earlier.
As described in detail in No. 1, for each pixel to be processed, an adjacent cell of Y 'data (4 bits) whose center defines a polygon containing the center of the image of the pixel,
The selected Y 'data is selected for each of the two types of adjacent LODs, and the selected Y' data is stored in the LOD (N) Y 'smoothing circuit 50 and the LOD (N +
M) It is supplied to the smoothing circuit 52. The smoothing circuits 50 and 52 generate LOD (N) and LOD (N + M), respectively, for the value (8 bits−8 bits−4 integers, 4 fractions) for the smoothed data Y _S (8 bits−4 bits).
4 are integers and 4 are fractions). Y 'mixing circuit 60
There LOD (N), LOD (N + M) ' receives _S, Y according to the procedure of appointment' smoothed data Y of combining _S data proportionally mixing Y _'B data (8-bit to 4-amino integer, four Fraction). Magnification circuit 70 receives a mixture Y _'B data, determines the Y data in response (16-bit) to it. Processing of the Y 'data is performed for each pixel in the span. The I 'and Q' data (4 bits each) are processed in the same manner as the Y 'data. However, the I 'data and the Q' data are processed for a grid having four equally spaced pixels per row or line and have a length of four equally spaced lines. Both the I 'and Q' grids cover the same area as the Y 'span. Each region of the processed I 'and Q' data covers an area of four adjacent pixels used to determine the Y 'data. The mixed value of each of the Y _B data corresponding to one of four adjacent pixels is
Associated with the same I 'and Q' data as described above.

Thus, a computer image generator for reducing the hardware and logic circuitry required to process color cell fabric information has been illustrated and described. Further, a computer image generator capable of using YIQ color processing has been described with reference to the drawings.

While certain preferred features of the invention have been illustrated by way of example,
Various modifications will occur to those skilled in the art. It is to be understood that the appended claims are intended to cover all such modifications that fall within the scope of the invention.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a normal RGB coordinate system and a corresponding YIQ system according to the present invention. FIG. 2 is a block diagram of an apparatus for obtaining YIQ data from RGB data according to the present invention. FIG. 4 is a schematic diagram of a Y data storage device for a predetermined area according to the present invention; FIG. 5 is an I and / or Q data for the same area as shown in FIG. 4 according to the present invention; 2 is a schematic diagram of a storage device. Explanation of main codes 30: RGBT database 35: Converter 38: Magnification circuit 40: Y'I'Q'T 'database

Continuation of the front page (56) References JP-A-57-4084 (JP, A) JP-A-62-92992 (JP, A) JP-A-62-296193 (JP, A) US Patent 4,905,164 (US, A) (58) Field surveyed (Int. Cl. ⁶ , DB name) G09G 5/00-5/40

Claims (12)

(57) [Claims] 1. A method for generating a color characteristic of an object in a scene to be displayed, which is used in a computer image generating apparatus, wherein a color in the scene is represented by a luminance component (Y), an in-phase chroma component (I) and a quadrature. Processing the Y component at a predetermined first fineness level to determine a composite Y component, represented by a phase chroma component (Q), and processing each of the I and Q components at a predetermined second fineness level Determining the composite I and Q components, respectively, wherein the second fineness level is coarser than the first fineness level and represents the color of the object in the scene where the composite Y, I and Q components are to be displayed. Determining translucency (T) data representing an object to be displayed in response to the composite Y component. 2. The method of claim 1 including the step of converting said composite Y, I and Q components into corresponding red (R), green (G) and blue (B) components. 3. The method of claim 1, wherein said second level of fineness is one fourth of said first level of fineness. 4. The method of claim 1, wherein the step of representing a color comprises selecting a Y, I, and Q component at a first bit resolution, and wherein the Y, I, and Q components are more than the first bit resolution. The Y, I and Q as represented by the lower second bit resolution
Multiplying each of the components by a magnification, processing the Y, I, and Q components of the second bit resolution to determine a combined Y, I, and Q component of the second bit resolution; The composite Y, I and Q components of the above are again multiplied by a factor to produce a composite Y, I and Q of the first bit resolution.
Forming the I and Q components and combining the first bit resolution Y, I
4. A method according to claim 3, including the step of causing the Q component to represent the color of the object in the scene to be displayed. 5. A method for generating a color characteristic of an object in a scene to be displayed, which is used in a computer image generating apparatus, wherein a color in the scene is represented by a luminance component (Y), an in-phase chroma component (I), and a quadrature. Processing the Y component at a predetermined first fineness level to determine a composite Y component, represented by a phase chroma component (Q), and processing each of the I and Q components at a predetermined second fineness level Determining the composite I and Q components, respectively, wherein the second fineness level is coarser than the first fineness level, and the composite Y, I and Q components represent the color of the object in the scene to be displayed. Wherein the translucency in the scene is represented by a translucency (T) component at a first bit resolution, and the T component is processed at a first fineness level to produce a composite at the first bit resolution. Determining a T component, multiplying the composite T component of the first bit resolution by a magnification,
Forming a composite T-component at a second bit resolution, wherein the first bit resolution is less than the second bit resolution. 6. A method for generating a color characteristic of an object of a scene to be displayed in a computer image generating apparatus, wherein red (R), green (G) and blue (B) color information representing the object to be displayed is provided. To corresponding luminance (Y), in-phase (I) chroma and quadrature (Q) chroma data, processing the Y, I and Q data to determine combined Y, I and Q data, Converting the combined Y, I, and Q data back into corresponding R, G, and B information so that the corresponding R, G, and B information represents the color of the object to be displayed; The Q data is represented at a first bit resolution, and the processing step comprises the following steps: the Y, I and Q data are represented at a second bit resolution lower than the first bit resolution. Multiplying the Y, I and Q data, processing the Y, I and Q data at the second bit resolution Determining the composite Y, I, and Q data of the second bit resolution, and multiplying again the composite Y, I, and Q data of the second bit resolution by the multiplication to produce the composite of the first bit resolution. Y, I
And Q data, and in response to the composite Y data, determining translucency (T) information representing an object to be displayed. 7. The processing step of: processing the Y data at a predetermined first level of fineness, and converting each of the I and Q data to a second predetermined level of coarseness that is greater than the first level of fineness. 7. The method of claim 6, comprising processing at a granularity level of: 8. The method of claim 7, wherein said second level of fineness is one-fourth of said first level of fineness. 9. An apparatus for generating color characteristics of an object of a scene to be displayed in a computer image generating apparatus, wherein a magnification is applied to first color information of an object represented by a first predetermined bit resolution. Multiplying an object represented by a second predetermined bit resolution lower than the first predetermined bit resolution
A first magnification means for generating color information of a third color of the object, the input being coupled to the first magnification means for receiving the second color information and representing the second bit resolution; Processing means for determining color information; coupled to the processing means for receiving third color information;
Second magnification means for generating, in response to the third color information, fourth color information of the object represented by the first bit resolution, wherein the fourth color information is to be displayed. A device that represents the color of 10. A computer image generating apparatus for generating a color characteristic of an object of a scene to be displayed, wherein a magnification is applied to first color information of the object represented by a first predetermined bit resolution. Multiplying an object represented by a second predetermined bit resolution lower than the first predetermined bit resolution
First magnification means for generating the color information of the first and second inputs; and an input coupled to the first magnification means,
Processing means for receiving the third color information of the object represented by the second bit resolution and determining the third color information of the object represented by the second bit resolution; receiving the third color information; Second magnification means for generating fourth color information of the object represented by the first bit resolution in response to the third color information, wherein the fourth color information is displayed. And the fourth color information is represented by luminance (Y), in-phase (I) chroma and quadrature (Q) chroma components, and the apparatus further comprises the fourth color information. And input means for determining red (R), green (G), and blue (B) color components corresponding to the supplied fourth color information; and The color information is represented by luminance (Y), in-phase (I) chroma and quadrature (Q) chroma components. Device, wherein they are coupled to the processing means, the Y component of the third color information includes a third magnification means for determining a translucency (T) information representative of the object to be display. 11. The second color information is represented by luminance (Y), in-phase (I) and quadrature (Q) chroma components, and the processing means processes the Y components at the first resolution. Y processing means, I processing means for processing the I component at the second resolution, and Q processing means for processing the Q component at the second resolution, wherein the first resolution is 11. A higher resolution than the second resolution.
The described device. 12. A computer image generating apparatus for generating color characteristics of an object of a scene to be displayed, wherein a magnification is applied to first color information of the object represented by a first predetermined bit resolution. Multiplying an object represented by a second predetermined bit resolution lower than the first predetermined bit resolution
First magnification means for generating the color information of the first and second inputs; and an input coupled to the first magnification means,
Processing means for receiving the third color information of the object represented by the second bit resolution and determining the third color information of the object represented by the second bit resolution; receiving the third color information; Second magnification means for generating, in response to third color information, fourth color information of the object represented by the first bit resolution, the fourth color information being displayed. And the first color information is represented by luminance (Y), in-phase (I) chroma and quadrature (Q) chroma components. Having a combined output and providing the first
Storage means for supplying color information of
And the corresponding I and Q components are stored at a second fineness level, and the first and second components are stored at the second fineness level.
Wherein the fineness level is higher than the second fineness level.